AVMER]ICAN MUSEUM NOVITATES PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CITY OF NEW YORK MAY 6, 1953 NUMBER 1626 THE SUBSPECIES OF OENEIS ALBERTA (LEPIDOPTERA, SATYRIDAE) By F. MARTIN BROWN' For some years Oeneis alberta oslari Skinner has been a very rare butterfly in collections. In early June, 1951, I had the good for- tune to find it rather plentiful in South Park, Park County, Colo- rado. The circumstances leading to its rediscovery and some notes on its habitat have been published (1952, Ent. News, vol. 63, pp. 119-123). This recently captured series of oslari affords the first opportunity to compare critically the three named entities that compose the species. The first-named of these, alberta Elwes and Edwards, is found on the prairies of Alberta and Saskatche- wan and is the best known of the three. During the following year Strecker described daura from the mountains of Arizona. This subspecies remained "lost" until D. K. Duncan, then of Globe, Arizona, recovered it in the vicinity of McNary in the White Mountains of Arizona during the early thirties. Oslari was last described. So far as I know the only true oslari in museum col- lections are the two pairs sent by Oslar to Skinner. All others that I have seen have been abberant Oeneis uhleri uhleri Reakirt. The data of Oslar's specimens misled collectors; they were taken in September, 1909. With almost 100 specimens representing the species before me it is quite evident that daura is distinctive a large, pale tawny creature. The other two subspecies are much alike, smaller and grayer than daura. In addition to size and color certain other items have been used to describe and differentiate the three butterflies: maculation, the limbal band on the upper side of the forewing, and the median band on the under side of the hind wing. These notes are based on my detailed study of the above variables. l Colorado Springs, Colorado. 2 AMERICAN MUSEUM NOVITATES NO. 1626 DESCRIPTIONS OF THE SAMPLES Oeneis alberta daura Strecker Chionobas daura STRECKER, 1894, Canadian Ent., vol. 26, p. 225. A series of 19 males and nine females collected by Duncan near the junction of Williams and Paradise creeks near the town of McNary in Apache County, Arizona, in meadows at an alti- tude of about 9000 feet, during the first nine days of June, 1933. So far as I can tell from the original description, my series is typical of the subspecies. Oeneis alberta alberta Elwes and Edwards Oeneis alberta ELWES AND EDWARDS, 1893, Trans. Ent. Soc. London, p. 467, pl. 15, fig. 2. A short series of eight males and two females collected by R. J. Fitch on the prairie near Lloydminster, Saskatchewan, Canada, in various years during the last three weeks of May. These butter- flies are smaller and darker than the material than I have seen from Alberta. A series of two males and eight females collected by C. Garrett at Carbon, Alberta, between June 5 and 18 which, with the follow- ing pair from Calgary, seem to be close to typical alberta as de- scribed. A male and a female from Calgary, Alberta, Canada. The first of these was collected by F. H. Dodd on May 23, 1894, and the second by C. Garrett on June 9. All the Alberta specimens are in the American Museum of Natural History. The Garrett material came from the Gunder Collection. Oeneis alberta oslari Skinner Chionobas alberta oslari SKINNER, 1911, Ent. News, vol. 22, p. 220. A series of 25 males and 14 females collected by me on June 11 and 12, 1951, on Crooked Creek and Trout Creek near Fairplay, Park County, Colorado, in meadows at an elevation of about 9800 feet. There is no doubt that this is the same insect as that described by Skinner. On June 4 and 5, 1952, I took 32 males and eight females at Crooked Creek. Of these a living pair was sent to Mr. Cyril F. dos Passos who has succeeded in bringing some of the larvae into the fifth instar. Mr. dos Passos reports that the larvae of alberta 1953 SUBSPECIES OF OENEIS ALBERTA 3 and oslari are very much alike. The native food plant is a species of Festuca, probably idahoensis. COLOR The use of descriptive terms for the colors observed in these insects does not lend itself to analysis. Instead I have arranged each series in an ascending order of color intensity, then divided each into a series of color categories that are recognizably different. On this basis the gamut of tones observed on both sexes falls into nine categories. The next step was to determine the per cent of each series that fell into each of the categories. The result of this was compared series-wise. Unfortunately I had returned the mate- rial borrowed from the American Museum of Natural History before doing this. Thus the distribution of color intensity reported CATEGORY 2 3 4 5 6 7 8 9 DAURA ( dO' OSLARI ...:.:.A ALBERTA d' DAURA QQ... ....... FIG. 1. The color range of the samples studied. Stippled blocks represent the range of the mid 50 per cent. Lines represent the total range on a nine-point scale. Based on table 1. TABLE 1 VARIATION~~~~~~~~~~~..............IN COLOR LSubsp~uspeciess CaLightestCategory Md5% CategoryDarkest Males daura 2 3.6-6.4 7 oslari 4 6.0-7.6 8 albertaa 6 7.2-8.0 9 Females daura 1 2.3-5.9 8 oslari 1 4,5-6.8 7 albertaa 8 a Based on nine males and two females from Saskatchewan. 4 AMERICAN MUSEUM NOVITATES NO . 1626 here for alberta applies only to my Saskatchewan series. My notes indicate that the Alberta material is somewhat lighter-"essenti- ally the same as that of oslari: possibly faded (?). " The results of this investigation are shown in table 1 and figure 1. It is evident that there is a strong tendency for the color to become darker from south to north in both sexes. This may be wholly a response to climatic conditions. The females tend to be a little lighter in color than the males. The two southern subspecies, daura and oslari, are quite distinct. On the other hand oslari does not differ appreciably from alberta when all of the Canadian specimens are considered. SIZE The measure of size that I have used is the radius of the left forewing taken with a vernier caliper to the nearest tenth of a e 11 AALBERLeA OSLARI% DAL/A 19 20 21 22 23 25 MILLIMETERS FIG. 2. The mean length of the left forewing. Circles have a radius of P.E On this and figures 3 to 7 the horizontal lines are spaced proportionally to the latitude differences. Based on table 2. millimeter. Size seems to decrease in both sexes from south to north. When the differences in size are divided by their probable 1953 SUBSPECIES OF OENEIS ALBERTA 5 errors, there appear to be statistically valid differences between daura and the other subspecies and between the males of oslari and alberta. I doubt that the difference in the latter case is of taxonomic importance. The parameters of size are given in table 2 and the statistics of the differences in table 3. Variation in size of the individuals within a series seems to be conservative. This may be a characteristic of Oeneis, since it is also TABLE 2 PARAMETERS OF SIZE Subspecies N Mean Radius S. E. V Males dauraa 19 22. 74 mm. 1. 59 mm. 7.0 oslari 25 20.53 0.81 3.9 alberta 11 19.38 0.71 3.6 Females daura 9 24.16 0.85 3.5 oslari 14 21.92 0.67 3.1 alberta 11 20.71 1.64 7.9 a If one 18.2-mm. dwarf is omitted, the figures are 18, 22.99, 1.07, and 4.8, respec- tively. TABLE 3 STATISTICS OF THE DIFFERENCES Pair Difference P. E. d P. E. Males daura-oslari 2.21 mm. 0.27 mm. 8.2 daura-alberta 3.36 0.29 11.6 oslari-alberta 1.15 0.19 6.0 Females daura-oslari 3.34 0.23 15.2 daura-alberta 3.45 0.40 8.6 oslari-alberta 1.21 0.37 3.3 true of uhleri. I have found generally a coefficient of variation around 7 to be normal for the size of butterflies. Among the in- sects here reported, Vis much less than this except for the males of daura and the females of alberta, where it is 7.0 and 7.9, respectively. It is possible that larger series will prove the lower numbers to be more normal for Oeneis. MACULATION There are partial series of spots in the limbal areas of both 6 AMERICAN MUSEUM NOVITATES NO. 1626 wings on both surfaces of these insects. For this study I tabulated those found on the upper side. Since asymmetry occurs I used only those on the left wings. In scoring each specimen I recorded these spots as "present," "traces," or "absent." Since no two persons would agree on the tallies for the first two categories I later lumped them together when preparing table 4. The trends shown on the upper side are repeated on the under side. Although there is a slight decrease in the development of the maculation from south to north, none of the differences found are statistically significant. This is especially true of the males. Among the females there are differences that do approach signifi- cance. Only on daura females does a spot occur with some regu- larity in M3-Cu1 on the hind wings.